Opening of tight junctions is involved in the pathogenesis of inflammatory bowel disease (IBD). Many studies have focused on the role of Tyr-kinases in the removal of tight junction components. However, no one has yet studied the role of atypical Protein Kinase C (aPKC), the evolutionarily conserved organizer of tight junctions, in inflammation. Our preliminary data shows a steep decrease of aPKC in intestinal epithelia exposed to pro- inflammatory signals. We have also shown that this effect is post-trancriptional and related to an Hsp-70- intermediate filament- dependent mechansim of rescue of aPKC. We hypothesize that post-translational mechanisms that recruit and maintain atypical PKC at the apical domain are impaired by inflammatory signals, resulting in increased tight junction permeability in the intestinal epithelium. We propose to: (1) Determine the molecular mechanisms by which Hsp70 chaperones rescue misfolded aPKC otherwise targeted for degradation. A combination of tissue culture of human intestinal cells, subcellular fractionation, and an in vitro reconstitution assay developed in our laboratory will be used to analyze the components of the aPKC rescue mechanisms. (2) Characterize molecular mechanisms that can antagonize loss of aPKC active conformation or degradation. Specific knock-down of Thr/Ser phosphatase catalytic subunits and expression of keratin intermediate filaments in human cells that normally lack them will be used to assess possible mechanisms to protect aPKC from losing its active conformation or to enhance the rescue mechanism. These observations will be validated in transgenic mice overexpressing keratin intermediate filaments. (3) Identify pro-inflammatory signals that downregulate aPKC and their targets in the degradation and rescue machinery that maintains aPKC physiologic levels. We will screen for pro-inflammatory cytokines that decrease aPKC levels, aside of TNF-, and analyze the step of the activation and degradation pathways of aPKC that are affected by pro-inflammatory signaling. Again, the results will be validated in an animal model